The long-term goal of chemotherapeutic pharmacogenomics research is to use genetic information to identify cancer patients at risk of particular drug-induced toxicities and adjust or alter their treatment accordingly. The goal of thi proposal is to discover and functionally validate genetic markers that predict treatment outcomes of the chemotherapeutic agent paclitaxel. Paclitaxel is a microtubule-stabilizing chemotherapy drug often used in the treatment of breast, ovarian and non-small-cell lung cancers, yet variable resistance and toxicities among individuals limit successful outcomes. Common paclitaxel-induced toxicities include peripheral sensory neuropathy and nephrotoxicity, which lead to discontinuation of therapy in approximately 20% of patients. Variation in drug response is likely to be dependent on the combined effects of variation in multiple gene products and environmental factors. By using a classic SNP-based genome-wide association (GWA) approach, the genetics of paclitaxel-induced cytotoxicity has been investigated in lymphoblastoid cell lines (LCLs) under a controlled environment. A preliminary analysis has shown that the top SNPs from a clinical trial of paclitaxel-induced peripheral neuropathy are enriched for SNPs associated with paclitaxel-induced cytotoxicity in LCLs, thus confirming a role for the LCL model in the analysis of genes involved in patient paclitaxel response. An overlap SNP from the preliminary analysis is located in an intron of RFX2. Decreased expression of this gene by siRNA resulted in increased sensitivity of NS-1 cells to paclitaxel measured by reduced neurite outgrowth and increased cytotoxicity, functionally validating the involvement of RFX2 in paclitaxel sensitivity. Although this approach was successful in identifying an important gene, our plan is to perform gene-based GWA studies because we realize that some of the associated genetic variants may not be detectable using a traditional SNP-based GWA approach. A novel gene-based approach to detect variants of small effect working together that incorporates expression quantitative trait loci (eQTLs), rare and common coding variants, and other functional information will increase power in a gene-level test for association with paclitaxel sensitivity. The fellow and research team also plan to compare the gene-based LCL results to those from clinical studies. Genes that overlap will make excellent candidates to test functionally in cell models of peripheral neuropathy. Additional genes identified in the computational analyses will be tested in the NS-1 model. Thus, in addition to identifying novel markers that contribute to variation in paclitaxel-induced phenotypes, these experiments will enhance understanding of the mechanisms involved in paclitaxel response. This work will provide a framework for choosing SNPs and genes that come from LCL GWA studies to interrogate in future clinical studies. By completing the aims of this proposal, the fellow will gain experience i human genetics and molecular biology while contributing to the field of translational cancer research. PUBLIC HEALTH RELEVANCE: Paclitaxel is a chemotherapy drug often used in the treatment of breast, ovarian and lung cancers, yet variable toxicities among individuals limit successful outcomes. If physicians could predict which patients are more likely to experience these severe side effects, lower doses or alternative treatments could be prescribed. The goal of this research is to discover and validate genetic markers that predict response to paclitaxel in order to improve outcomes for cancer patients.